Slave - Exception: Failed to update slaveInfo:

Deadline
1

Hi,

I am noticing this error appearing on our deadline slaces

Slave - Exception: Failed to update slaveInfo: The process cannot access the file because it is being used by another process. Slave - No slave update error notification address specified in Repository Options - cannot send notification

What does this mean and what issues might it cause.

Everything seems to be rendering fine still.

Using Deadline 4.1.0.4270

Many thanks,

Richard

2

It means that the slave couldn’t update its state, which is stored in the .slaveInfo file in \your\repository\slaves[SLAVENAME]. The slave tries to update its state every 20-40 seconds, so it’s not a problem if it happens occasionally. If it happens a lot (due to a network or permissions issue), then the slave could end up appearing stalled in the Monitor.

Cheers,

  • Ryan
3

Well it is happening every couple of minutes while the slave is idle.

The slaves start rendering fine when a job is submitted though.

Not sure why there would be permission issues. All slaves can see the mapped drives no problem.

I have left the studio now but will check on monday regarding the permissions on the .slaveInfo file and post any findings.

Many thanks,

Richard

4

Cool. I should note that we improved on this a bit in 5.1. Before, it would only make one attempt to write before it complained. Now it makes 3 (with a slight delay between them). Since it only happens every few minutes, it’s probably not permission related. Maybe the repository server is holding on to its file handle locks longer than it should…

5

I checked the .slaveinfo file on all the slaves and the permissions are full read and write for all users so would like to think it would not be a permissions issue.

I think we will upgrade to 5.1 but a bit worried about upgrading and causing problems. At the moment everything renders fine. (if it aint broke don’t fox it mentality)

I think the upgrading section of the docs is a little vague.

Do I have to delete the 4.1 repository? Or with 5.1 do I just create a new repository?

Then I just uninstall the 4.1 slaves/clients etc from the workstations and render nodes and install 5.1?

Thanks,

Richard

6

The Exception error is happening every minute or so.

7

You just need to create a new 5.1 repository. You don’t have to delete the original. I would recommend creating the new 5.1 repository first, and then slowly start migrating nodes from 4.1 to 5.1. This way, your 4.1 farm can still be active during the migration.

For the clients, it’s definitely not a bad idea to uninstall 4.1 first, but it’s not required. You will have to run the 5.1 client installer on each client machine.

Cheers,

  • Ryan
8

Ok thanks Ryan.

I will start doing this. I first need to request a new license from the sales team for 5.1 as it says it uses a different license to previous versions.

Best,

Richard